INDEX

Explanations

references to specific times, especially months of the year and words like 'last' and 'earlier'

oai_token-act-pair · gemini-2.0-flash

year/seasons

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_7/width_16k/average_l0_70

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 transfieras

-0.65

DockStyle

-0.63

friday

-0.57

 monday

-0.56

 Thursday

-0.56

 Tuesday

-0.55

Yesterday

-0.54

帖最后由

-0.54

̍t

-0.54

 Monday

-0.53

POSITIVE LOGITS

RetentionPolicy

0.61

 honnête

0.59

 året

0.58

شهاد

0.56

揄

0.55

TagMode

0.54

 Frühjahr

0.54

 kış

0.53

DoubleQuotes

0.53

 year

0.53

Activations Density 2.820%

references to specific times, especially months of the year and words like 'last' and 'earlier'

year/seasons

No Comments

No Known Activations